WO2013050745A1 - Connection method - Google Patents
Connection method Download PDFInfo
- Publication number
- WO2013050745A1 WO2013050745A1 PCT/GB2012/052413 GB2012052413W WO2013050745A1 WO 2013050745 A1 WO2013050745 A1 WO 2013050745A1 GB 2012052413 W GB2012052413 W GB 2012052413W WO 2013050745 A1 WO2013050745 A1 WO 2013050745A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- data
- failure
- communication
- devices
- server
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
- G06F11/0751—Error or fault detection not based on redundancy
- G06F11/0754—Error or fault detection not based on redundancy by exceeding limits
- G06F11/0757—Error or fault detection not based on redundancy by exceeding limits by exceeding a time limit, i.e. time-out, e.g. watchdogs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
- G06F11/0706—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation the processing taking place on a specific hardware platform or in a specific software environment
- G06F11/0745—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation the processing taking place on a specific hardware platform or in a specific software environment in an input/output transactions management context
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
- G06F11/0751—Error or fault detection not based on redundancy
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
- G06F11/0793—Remedial or corrective actions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
- G06F13/38—Information transfer, e.g. on bus
- G06F13/42—Bus transfer protocol, e.g. handshake; Synchronisation
- G06F13/4204—Bus transfer protocol, e.g. handshake; Synchronisation on a parallel bus
- G06F13/4221—Bus transfer protocol, e.g. handshake; Synchronisation on a parallel bus being an input/output bus, e.g. ISA bus, EISA bus, PCI bus, SCSI bus
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
- G06F11/0766—Error or fault reporting or storing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/16—Error detection or correction of the data by redundancy in hardware
- G06F11/20—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
- G06F11/2002—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where interconnections or communication control functionality are redundant
- G06F11/2005—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where interconnections or communication control functionality are redundant using redundant communication controllers
Definitions
- the present invention relates to a method of managing a failure of communication between a first device and a second device.
- a processing device it is often necessary to connect a processing device to a plurality of input and output devices.
- Appropriate data communication is achieved by connecting the devices in such a way as to allow them to send data packets to each other over a link, which may be a wired link or a wireless link.
- PCI Express for example, was used only for permanent connection of I/O adaptors within the server (i.e. "in-box").
- I/O adaptors within the server
- failure of the I/O device represents a failure of the server as a whole and as such, continued operation of that server is not considered to be necessary.
- An increasing interest in remotely connected PCI Express I/O devices i.e. I/O devices connected outside a server
- the sending and receiving of data packets is often described in terms of transactions. A transaction involves one or more data packets being sent between devices.
- PCI Express implements a split transaction model, wherein a source device transmits a request data packet to a destination device, and awaits a completion data packet from the destination device in response.
- operating systems are not adapted to handle failed PCI Express transactions gracefully. For example, if a server sends a request data packet to a connected device and, unexpectedly, receives no completion data packets in response to that request, the operating system of the server is likely to crash. As such, current connected systems based on PCI Express are likely to crash when a connected PCI Express resource becomes unexpectedly unavailable.
- a method of managing a failure of communication between a first device and a second device comprising: detecting at a third device intermediate the first and second devices failure of communication between first device and the second device; transmitting first data from the third device to the first device, the first data indicating that the second device is unusable by the first device.
- the third device can send appropriate data to the first device to prevent failure of the first device.
- the first data may hide the failure of communication from software (OS, drivers, applications) operating on the first device, by informing the first device that the second device has changed status.
- the third device may emulate the continued existence of the second device, but in a state which the first device interprets as "available, but not usable".
- the first device may be adapted to gracefully handle the "available, but not usable" state without service interruption (for example by communicating with an alternative to the second device).
- the third device may be within a failure unit defined by the first device, while the second device may be outside the failure unit defined by the first device.
- the first data may indicate that the second device is locally functional but unable to perform its designated service due to an external problem, which may be temporary.
- the method may further comprise intercepting, at the third device, second data from the first device to the second device.
- the third device may be between the first and second devices such that communication between the first and second devices is via the third device.
- the third device may take the form of a cable adaptor.
- data sent from the first device to the second device may be re-directed to the third device by a further device between first and second devices.
- Detecting failure of communication between the first and second devices may comprise detecting that a response to the second data will not be received by the first device.
- Detecting failure of communication between the first and second devices may comprise detecting that a cable connecting the first and second devices has been disconnected.
- detecting a cable disconnect may comprise detecting the absence of a previously present "cable detect" signal.
- detecting failure of the communication may comprise detecting an interruption or interference in the wireless connection.
- a part of the first data intended to indicate a source of the first data may indicate the second device.
- the first data may comprise data indicating that the first data is in response to the second data.
- the second data may comprise a data field indicating a transaction to which the second data belongs and the first data may comprise data indicating that it belongs to the same transaction.
- One of the first and second devices may be a server, and/or one of the first and second devices may be a remotely connected resource of the first or second device.
- one of the first and second devices may be an I/O device or other remote resource.
- the first device may be connected to the second device via a PCI Express connection.
- the first data may indicate that the second device is unreachable, the first data may indicate that the second device has malfunctioned, and/or the first data may indicate that the first data is corrupt.
- the first data may comprise a status bit having a value of "1 ", the status bit indicating that the second device has malfunctioned and/or that the first data is corrupt.
- Detecting failure of communication between the first and second devices may comprise transmitting third data from the third device to the second device, and if a response to the third data is not received from the second device within a predetermined time period determining that failure of communication between the first and second devices has occurred.
- an apparatus for managing a failure of communication between a first device and a second device comprising: means for detecting at a third device intermediate the first and second devices failure of communication between first device and the second device; means for transmitting first data from the third device to the first device, the first data indicating that the second device is unusable by the first device.
- a resilient cable adaptor for connection between a first device and a second device, comprising: a detector arranged to detect failure of communication between a connected first and second device; a transmitter arranged to transmit first data to a connected first device, the first data indicating that a connected second device is unusable by the first device.
- FIGS. 1 and 1 a are schematic illustrations of a prior art I/O configuration in which I/O devices are provided as components of a server;
- FIG. 2 is a schematic illustration of a prior art I/O configuration in which an I/O device is remotely connected to a server;
- Figure 3 is a schematic illustration of a prior art I/O configuration in which a plurality of I/O devices are remotely connected to a server;
- Figure 4 is a schematic illustration of an I/O configuration in which an I/O device is connected to a server according to an embodiment of the present invention
- Figure 5 is a schematic illustration of a data packet header
- FIG. 6 is a schematic illustration of an I/O configuration in which a server is connected to an I/O device according to at embodiment of the present invention.
- a server 1 comprises a CPU/PCI Express Root Complex (CPU/RC) 2 and a Network Interface Controller (NIC) 3.
- the CPU/RC 2 is connected to the NIC 3 via a PCI Express chip-to-chip connection 4.
- the NIC 3 connects to a network 5 via an Ethernet connection 6 (which may be, for example, a cable or a wireless connection).
- the server 1 is configured to provide a user of the server 1 with access to the network 5. It will be appreciated that other details of the server 1 not pertinent to the present invention have been omitted for clarity, those details being readily apparent to those skilled in the art.
- the server 1 of Figure 1 is considered to be a single failure unit. Failure of an internal component of the server 1 (such as the NIC 3 or the connection 4) is considered to be a failure of the whole of the server 1 , as the server 1 can no longer provide its intended function (i.e. facilitation of access to the network 5). As such, upon failure of the NIC 3 or the communication between the NIC 3 and the CPU/RC 2, there is no requirement for the server 1 to continue functioning and the server 1 will fail.
- an internal component of the server 1 such as the NIC 3 or the connection 4
- the failure of an external component i.e. outside the single failure unit of the server 1
- an external component i.e. outside the single failure unit of the server 1
- the Ethernet connection 6 or a downstream switch (not shown)
- failure of an externally connected component of the server 1 should not result in failure of the server 1 given that replacement devices may be easily connected (or re-connected in the event, for example, of unintentional disconnection).
- the server 1 is therefore adapted to gracefully handle situations where it is properly informed that external resources are currently unavailable.
- failure of the Ethernet connection 6 causes the NIC 3 to inform the CPU/RC 2 that the Ethernet connection 6 is unavailable (by transmission of a suitable error message).
- the CPU/RC 2 can cease sending data packets to the NIC 3 for transmission over the Ethernet connection 6, without causing failure of the server 1 as a whole.
- a CPU/RC may be able to continue to provide its designated service by utilising backup resources where available.
- a server 1 a comprises a CPU/RC 2a connected to a NIC 3a via a PCI Express chip-to-chip connection 4a and to a NIC 3b via a PCI Express chip-to-chip connection 4b.
- Each of the NICs 3a, 3b connect to the network 5 via respective Ethernet connections 6a, 6b. As such, each of the NICs 3a, 3b can facilitate access to the network 5.
- the corresponding NIC 3b Upon failure of the Ethernet connection 6b, for example, the corresponding NIC 3b transmits an error message to the CPU/RC 2a indicating that the Ethernet connection 6b is unavailable. As the CPU/RC 2a is adapted to gracefully handle such error messages, the CPU/RC 2a can continue to provide its designated function using the NIC 3a and the Ethernet connection 6a.
- a NIC 7 is housed externally to a server 8 (i.e. outside the single failure unit defined by the server 8).
- the server 8 comprises a PCI Express cable adaptor 9 connected to a CPU/RC 10 via a PCI Express connection 1 1 .
- the PCI Express cable adaptor 9 connects the CPU/RC 10 to a PCI Express cable adaptor 12 housed within a remote I/O appliance 13 via a cable 14.
- the remote I/O appliance 13 also houses the NIC 7, which is connected by a PCI Express connection 15 to the PCI Express cable adaptor 12.
- the PCI Express cable adaptors 9, 12 provide only cable-driving and signal conditioning functionality required to transport PCI Express signals over the cable 14. That is, the cable adaptors 9, 12 do not provide any logical functions, such that, to (software operating on) the server 8, the system of Figure 2 is logically identical to that of Figure 1 .
- failure of the cable 14, or failure of the NIC 7 itself severs communication between the CPU/RC 10 and the NIC 7. As such, the CPU/RC 10 will receive no response to outstanding transactions with the NIC 7. As described above, the lack of response to an outstanding transaction results in failure of the server 8 as a whole. While in the arrangement of Figure 1 , it is generally acceptable that failure of the NIC 3 (or the connection 4 between the NIC 3 and the CPU/RC 2) constitutes failure of the server 1 as a whole, the remote nature of the cable 14 and NIC 7 in the arrangement of Figure 2 means that these components can be easily replaced or re-connected. As such, in the arrangement of Figure 2, it is undesirable that failure, either of the NIC 7 or the cable 14, should cause failure of the server 8.
- a CPU/RC 19 of a server 20 is connected, via a cable adaptor 21 , to a remote I/O appliance 22 via a cable 23 and to a remote I/O appliance 24 via a cable 25.
- the remote I/O appliance 22 comprises a cable adaptor 26, for connection to the cable 23, and a NIC 27 connected to the cable adapter 26 via a connection 28.
- the remote I/O appliance 24 comprises a cable adapter 29, for connection to the cable 25, and a NIC 30 connected to the cable adapter 29 via a connection 31 .
- Each NIC 27, 30 is connected to the network 5 via respective connections 32, 33.
- the server 20 may continue to provide its designated service (i.e. connection to the network 5) using the other of the I/O appliances 22, 24.
- the CPU/RC 19 will receive no response to data packets transmitted to a disconnected one of the NICs 27, 30 resulting in an undesirable failure of the server 20.
- Figure 4 illustrates the general arrangement of Figure 2, modified according to an embodiment of the present invention.
- a cable adaptor within the failure unit defined by a server is replaced with a resilient cable adaptor, the operation of which is described in more detail below.
- the resilient cable adaptor converts fatal errors (such as the failure of a PCI cable) into non fatal errors.
- a server 35 comprises a CPU/RC 36 connected to a resilient cable adaptor 37 via a PCI Express connection 38.
- the resilient cable adaptor 37 connects the CPU/RC 36 to a PCI Express cable adaptor 38 housed within a remote I/O appliance 39 via a cable 40.
- the remote I/O appliance 39 houses a NIC 41 , which is connected by a PCI Express connection 42 to the PCI Express cable adaptor 38, and to the network 5 via an Ethernet connection 43.
- the resilient cable adaptor 37 provides logic and hardware necessary to monitor for failure of components downstream of the resilient cable adaptor 37 (for example the cable 40, the NIC 41 , etc).
- the resilient cable adaptor 37 is configured to detect when a cable is unplugged (for example, by way of a provided "presence detect" signal, or any other appropriate method) and to listen for messages from the remote I/O appliance 39 indicating a problem with the NIC 41 .
- the resilient cable adaptor 37 is further adapted to inspect data packets in order to determine and record information regarding transactions initiated by the server 35, and to await the appropriate responses.
- the resilient cable adaptor 37 can wait for a predetermined time period, the expiry of which constitutes the occurrence of a failure. For example, a threshold for the time period may be set such that it expires earlier than a time at which the server 35 would crash as a result of not receiving a response data packet.
- the resilient cable adaptor 37 may also be configured to generate data packets independently of the CPU/RC 36 for transmission to the NIC 41 and to await an appropriate response (whereby lack of a response within a predetermined time period constitutes failure of the NIC 41 ). It will be appreciated that failure of components downstream of the resilient cable adaptor 37 may have one or more of a plurality of differing causes, and that the resilient cable adaptor 37 may implement any appropriate means for detecting such events as will be readily apparent to those skilled in the art.
- the resilient cable adaptor 37 Upon detection of the failure of a downstream component, the resilient cable adaptor 37 emulates the NIC 41 to provide the CPU/RC 36 with a suitable non-fatal error message.
- the non-fatal error message indicates that the NIC 41 is available but in a state in which it cannot be used (for example, emulating a state in which the Ethernet cable 43 has been unplugged).
- the resilient cable adaptor 37 may generate data packets which appear to have been generated by the NIC 41 .
- the NIC 41 may have a plurality of independent device functions, a maximum of eight functions being supported by the PCI Express protocol. That is, the NIC 41 may appear to the CPU/RC 36 to be up to eight separate devices. Each device function of the NIC 41 has a corresponding identifier uniquely identifying that function. Data packets sent from a particular device function of the NIC 41 have a transaction identifier comprising a requester identifier that corresponds to the identifier of the device function sending the data packet.
- a requester identifier 45 identifies a device function which originated the data packet and comprises sixteen bits indexed from 0 to 15. It can be seen that the requester identifier 45 comprises a bus number field 46 occupying an upper eight bits, device number field 47 occupying a central five bits and function number field 48 occupying a lower three bits.
- bus number 46 occupying an upper eight bits
- device number field 47 occupying a central five bits
- function number field 48 occupying a lower three bits.
- the packet header shown in Figure 5 further comprises a tag field 49 comprising eight bits.
- a transaction may be made up of a request data packet and one or more corresponding completion data packets. Each request data packet is associated with a value which is stored in the tag field 49. Each corresponding completion data packet has the same value stored in the tag field 49, thus associating a completion data packet with the relevant request data packet.
- Unique tag values are allocated to all outstanding requests that require one or more completion data packets from a destination device. Given that the tag field 49 has eight bits, two hundred and fifty-six (2 8 ) possible tag values can be represented.
- the function number field 48 is provided with a function number of the relevant function of the device sending a request. If a device has fewer than eight functions, there may be unused bits in the function number field 48. It is therefore known to use only sufficient bits of the function number field 48 to represent the functions of a device and to use any unused bits of the function number field 48 as a phantom function number which is logically combined with the tag field 29. Where only a single function is provided all bits of the function number field 48 can be logically combined with the tag field 49 to provide support for up to two thousand and forty-eight (2 11 ) outstanding requests.
- the resilient cable adaptor 37 can examine a received data packet, and in particular the tag field and requester ID, in order to determine the transaction identifier of that data packet. This data can be stored in on-board memory, provided by the resilient cable adaptor. Data packets created by the resilient cable adaptor 37 can then be provided with the correct transaction identifier (if the type of error message being provided requires a transaction ID) such that the CPU/RC 36 believes that the messages relate to an outstanding transaction between the CPU/RC 36 and the NIC 41 . Similarly, data packets can be generated by the resilient cable adaptor 37 having the requester ID of the NIC 41 , so that those data packets appear to have been generated by the NIC 41 .
- a suitable error message is chosen in dependence upon the information recorded by the resilient cable adaptor 37 about the transaction. For example, where the CPU/RC 36 transmitted a register read request data packet to the NIC 41 , upon failure of a downstream component severing communication between the CPU/RC 36 and the NIC 41 , the resilient cable adaptor 37 may provide a data packet in which an error status bit is set to a value of "1 ". Software operating on the server 1 will interpret receipt of the data packet in which the status bit is set to indicate that an error has occurred, but the error will not cause failure of the server 35. As a further example, if the transaction concerns data transmitted over the network 5, the resilient cable adaptor 37 may transmit a message to the CPU/RC 36 indicating that the Ethernet connection 43 has been disconnected. Each error message may be hard coded into the onboard memory of the resilient cable adaptor 37. Logic operating on the resilient cable adaptor 37 may then be adapted to determine which of the stored messages is appropriate in light of a transaction which is awaiting a completion data packet.
- error message provided to the CPU/RC 36 may further depend upon the particulars of the remotely connected device which the resilient cable adaptor 37 is emulating. For example, where the remotely connected device is a storage resource, error messages indicating that an Ethernet connection has been severed would be inappropriate. As such, the error messages may be tailored to the type of remotely connected device, to indicate that that device is locally functional but unable to perform its designate service (i.e. to replace a fatal error with a non-fatal error).
- any device remotely connected to the server 35 has a state in which the server 35 considers the device to be unusable but which does not cause the server 35 to crash; and second, that the unusable state is simple compared to the fully operational states of the I/O device.
- a remotely connected I/O device fails when a connected server (or a connection to a server) fails.
- This is particularly, but not exclusively, appropriate in systems where a remote I/O device connects to, and is shared by, a plurality of servers (for example, in systems implementing multi-root I/O virtualization). It will be appreciated that the failure of one server should not cause the remote I/O device to fail and hence be unable to provide its service to the remaining server(s).
- Figure 6 illustrates the general arrangement of Figure 2, modified according to an embodiment of the present invention.
- a cable adaptor within a remote I/O appliance is replaced with a resilient cable adaptor.
- a server 50 comprises a CPU/RC 51 connected to a cable adaptor 52 via a PCI Express connection 53.
- the cable adaptor 52 connects the CPU/RC 51 to a resilient cable adaptor 54 housed within a remote I/O appliance 55 via a cable 56.
- the remote I/O appliance 55 also houses a NIC 57, which is connected by a PCI Express connection 58 to the resilient cable adaptor 54, and to the network 5 via an Ethernet connection 59.
- the resilient cable adaptor 54 is adapted to detect failures of components upstream of the resilient cable adaptor 54. Upon detection of an upstream component failure, the resilient cable adaptor 54 is adapted to emulate the existence of the server 50.
- the resilient cable adaptor 54 may be configured to issue completion data packets in response to outstanding memory read/write requests issued by the NIC 57, in a timely fashion, with an non-fatal error status.
- the error messages provided to the NIC 57 will depend upon the transaction for which a completion data packet is required.
- configurations of servers and remote resources may include both the resilient cable adaptor 37 and the resilient cable adaptor 54. That is, in some configurations a resilient cable adaptor may be provided both within one or more respective failure units defined by one or more servers, and within one or more failure units defined by one or more respective remotely connected resources.
- NICs PCI Express Ethernet network interface cards
- server PCI Express Ethernet network interface cards
- the present invention is more generally applicable. Indeed, the present invention may be used in systems utilising interconnects other than PCI Express, and with remote devices other than NICs (for example, Fibre Channel Host Bus Adaptors, storage controllers etc).
- systems utilising the present invention may comprise a plurality of servers each having connections to a plurality of independent remote devices. Such arrangements would allow each server to continue useful operation in the event of a failure associated with one remote device by using another remotely connected device.
- the resilient cable adaptor 37, 54 maintains a record of which server or which remote resource has failed in order to select an appropriate non-fatal error message (taking into account the active transactions of the failed server or remote resource where necessary).
- resilient cable adaptor is described above as a single device comprising the functionality of the present invention and the functionality of a standard cable adaptor, the present invention may be implemented in a dedicated device.
- a device implementing the present invention may be placed between the CPU/RC 10 and the cable adaptor 9, or between cable adaptor 9 and the cable 14.
- the resilient cable adaptor 37, 54 may be implemented in a Field Programmable Gate Array (FPGA) or an Application-specific Integrated Circuit (ASIC). It will be appreciated, however, that the resilient cable adaptor 37, 54 may be implemented using any suitable means.
- FPGA Field Programmable Gate Array
- ASIC Application-specific Integrated Circuit
- server is intended broadly and is intended to cover any computing device.
Abstract
Description
Claims
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020147011657A KR102018225B1 (en) | 2011-10-05 | 2012-09-28 | Connection Method |
JP2014533980A JP6134720B2 (en) | 2011-10-05 | 2012-09-28 | Connection method |
EP12787480.8A EP2764438B1 (en) | 2011-10-05 | 2012-09-28 | Connection method |
CN201280048841.9A CN103858105B (en) | 2011-10-05 | 2012-09-28 | connection method |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
GB1117230.1A GB2495313B (en) | 2011-10-05 | 2011-10-05 | Connection method |
GB1117230.1 | 2011-10-05 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2013050745A1 true WO2013050745A1 (en) | 2013-04-11 |
Family
ID=45035231
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/GB2012/052413 WO2013050745A1 (en) | 2011-10-05 | 2012-09-28 | Connection method |
Country Status (7)
Country | Link |
---|---|
US (1) | US9798601B2 (en) |
EP (1) | EP2764438B1 (en) |
JP (1) | JP6134720B2 (en) |
KR (1) | KR102018225B1 (en) |
CN (1) | CN103858105B (en) |
GB (1) | GB2495313B (en) |
WO (1) | WO2013050745A1 (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140365699A1 (en) * | 2013-06-11 | 2014-12-11 | Allied Telesis Holdings Kabushiki Kaisha | Adapter card for thin computing devices |
US10466220B1 (en) | 2018-09-21 | 2019-11-05 | Pace Analytical Services, LLC | Alerting for instruments that transfer physical samples |
KR102166729B1 (en) | 2018-12-19 | 2020-10-19 | 주식회사 포스코 | Coil tilting device |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5864653A (en) * | 1996-12-31 | 1999-01-26 | Compaq Computer Corporation | PCI hot spare capability for failed components |
US20060136789A1 (en) * | 2004-12-16 | 2006-06-22 | Nec Corporation | Computer system, method of detecting a stall in a computer system, and signal-bearing medium embodying a program causing a computer system to perform a method of detecting a stall in a computer system |
Family Cites Families (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR100193809B1 (en) * | 1995-11-16 | 1999-06-15 | 윤종용 | How to notify error status in fax |
US20020194319A1 (en) * | 2001-06-13 | 2002-12-19 | Ritche Scott D. | Automated operations and service monitoring system for distributed computer networks |
JP2005172218A (en) * | 2003-11-21 | 2005-06-30 | Hama International:Kk | Pipe joint |
JP4416563B2 (en) * | 2004-04-19 | 2010-02-17 | キヤノン株式会社 | Network device management apparatus, control method therefor, computer program, and computer-readable storage medium |
JP4328672B2 (en) * | 2004-06-01 | 2009-09-09 | キヤノン株式会社 | Information processing apparatus and device |
JP2007094706A (en) * | 2005-09-28 | 2007-04-12 | Konica Minolta Business Technologies Inc | Data processor, system and method for coping with cable connection abnormality |
US20070240209A1 (en) * | 2006-04-05 | 2007-10-11 | Lewis Adam C | Session persistence on a wireless network |
JP4369470B2 (en) * | 2006-12-26 | 2009-11-18 | 富士通株式会社 | Data relay apparatus, storage apparatus, and response delay monitoring method |
US20080288633A1 (en) * | 2007-05-18 | 2008-11-20 | Kestrelink Corporation | Network device tracking using a network device server |
JP5151500B2 (en) * | 2008-01-18 | 2013-02-27 | 日本電気株式会社 | Computer system, failure processing method, and failure processing program |
GB2460014B (en) * | 2008-04-28 | 2011-11-23 | Virtensys Ltd | Method of processing data packets |
US7757014B2 (en) * | 2008-05-14 | 2010-07-13 | Tsung-Hsien Ho | Method for disconnecting a transceiver from a bus in multipoint/multidrop architecture |
WO2010071628A1 (en) * | 2008-12-15 | 2010-06-24 | Hewlett-Packard Development Company, L.P. | Detecting an unreliable link in a computer system |
JP5281942B2 (en) * | 2009-03-26 | 2013-09-04 | 株式会社日立製作所 | Computer and its fault handling method |
JP2010238150A (en) * | 2009-03-31 | 2010-10-21 | Toshiba Corp | Pciexpress communication system and communication method thereof |
JP2010288148A (en) * | 2009-06-12 | 2010-12-24 | Ricoh Co Ltd | Data communication apparatus, data communication control method, data communication control program, and recording medium |
US20110047413A1 (en) * | 2009-08-20 | 2011-02-24 | Mcgill Robert E | Methods and devices for detecting service failures and maintaining computing services using a resilient intelligent client computer |
-
2011
- 2011-10-05 GB GB1117230.1A patent/GB2495313B/en active Active
- 2011-10-17 US US13/275,296 patent/US9798601B2/en active Active
-
2012
- 2012-09-28 EP EP12787480.8A patent/EP2764438B1/en active Active
- 2012-09-28 WO PCT/GB2012/052413 patent/WO2013050745A1/en active Application Filing
- 2012-09-28 CN CN201280048841.9A patent/CN103858105B/en active Active
- 2012-09-28 KR KR1020147011657A patent/KR102018225B1/en active IP Right Grant
- 2012-09-28 JP JP2014533980A patent/JP6134720B2/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5864653A (en) * | 1996-12-31 | 1999-01-26 | Compaq Computer Corporation | PCI hot spare capability for failed components |
US20060136789A1 (en) * | 2004-12-16 | 2006-06-22 | Nec Corporation | Computer system, method of detecting a stall in a computer system, and signal-bearing medium embodying a program causing a computer system to perform a method of detecting a stall in a computer system |
Also Published As
Publication number | Publication date |
---|---|
US9798601B2 (en) | 2017-10-24 |
JP2014532236A (en) | 2014-12-04 |
CN103858105B (en) | 2017-08-29 |
EP2764438B1 (en) | 2019-04-24 |
KR20140090171A (en) | 2014-07-16 |
US20130091271A1 (en) | 2013-04-11 |
CN103858105A (en) | 2014-06-11 |
GB2495313B (en) | 2013-12-04 |
GB201117230D0 (en) | 2011-11-16 |
GB2495313A (en) | 2013-04-10 |
JP6134720B2 (en) | 2017-05-24 |
EP2764438A1 (en) | 2014-08-13 |
KR102018225B1 (en) | 2019-09-05 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP2052326B1 (en) | Fault-isolating sas expander | |
US8892936B2 (en) | Cluster wide consistent detection of interconnect failures | |
US7536584B2 (en) | Fault-isolating SAS expander | |
WO2021027481A1 (en) | Fault processing method, apparatus, computer device, storage medium and storage system | |
US20080162984A1 (en) | Method and apparatus for hardware assisted takeover | |
JP2017518552A (en) | High-speed serial link in-band lane failover for RAS and power management | |
JPH09508483A (en) | Networking module and method for fault tolerant transmission of system management information | |
US7774638B1 (en) | Uncorrectable data error containment systems and methods | |
JPH086910A (en) | Cluster type computer system | |
JP6183931B2 (en) | Cluster system, server apparatus, cluster system management method, and program | |
EP2764438B1 (en) | Connection method | |
WO1999021322A9 (en) | Method and system for fault-tolerant network connection switchover | |
US9760460B2 (en) | Storage system, storage device, and monitoring server | |
US10235045B2 (en) | Storage system and control apparatus | |
GB2267372A (en) | Fault detection in an information handling system | |
JPH07168790A (en) | Information processor | |
CN103023697B (en) | A kind of management method, Apparatus and system of array multipath | |
US8050176B2 (en) | Methods and systems for a data processing system having radiation tolerant bus | |
US11669399B2 (en) | System and method for fault identification and fault handling in a multiport power sourcing device | |
JP2004007930A (en) | System and program for controlling power system monitoring | |
US10599510B2 (en) | Computer system and error isolation method | |
EP4174668A1 (en) | Communication interface adapter, method for dynamic pid assignment, and method for automatically diagnosing peripheral device and recovering peripheral device found to be in abnormal operation | |
KR100771915B1 (en) | Digital protection relay and communication duplex method | |
CN117851300A (en) | Multi-AZ arbitration system and method | |
WO2017199288A1 (en) | Computer system and failure management method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 12787480 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2012787480 Country of ref document: EP |
|
ENP | Entry into the national phase |
Ref document number: 2014533980 Country of ref document: JP Kind code of ref document: A |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
ENP | Entry into the national phase |
Ref document number: 20147011657 Country of ref document: KR Kind code of ref document: A |